INDEX

Explanations

phrases that indicate details or distinct characteristics, emphasizing specificity in content

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ony

-0.17

 mere

-0.17

Boy

-0.16

à¸£

-0.15

ish

-0.15

hol

-0.15

oric

-0.15

boys

-0.15

 entire

-0.15

iesel

-0.14

POSITIVE LOGITS

ities

0.20

-purpose

0.20

ially

0.19

 biá»ĩt

0.18

 sayÄ±da

0.16

blr

0.15

ÑĮ

0.15

idades

0.15

ummings

0.15

ulty

0.15

Activations Density 0.037%

phrases that indicate details or distinct characteristics, emphasizing specificity in content

No Comments

No Known Activations